And scores don't matter, they are just arbitrary numbers that have no correlation to performance
Sent from my Nexus 7 using Tapatalk 4 Beta
I'll add to my previous post to give a little better context. This debate has been hashed out many times here and elsewhere but I think at this point it's pretty much accepted that scores aren't an accurate indicator of real performance.
Scores are useful in so far as they'll give you an idea of how a single change to the same kernel might have affected the tests the app does to produce the score.
That difference in the score from that change may or may not translate into an "improvement" in your experience with the device. Add to the fact that there's so much that goes into a kernel and that when you start comparing scores across multiple kernels with very different configurations, the scores begin to have really very little value in how well the kernel performs in real world settings.
All that is just to point out that having a higher score doesn't make the kernel (or a ROM) better. Most devs gave up trying to achieve high benchmarks a while ago. Users seem to want some single data point that says something is better but the reality is that it's more complicated that and the only way to determine what kernel is best for you is to try them all and weigh their benefits as a whole.
Franco tends to be one of the more well rounded kernel's out there and has a great mix of performance, battery life, control and features. As previously suggested, flip the power mode over to PowerfulPerformance and I'd bet the benchmark score jumps up. By default Franco is rigged for a very balanced experience.
Hope that helps a bit more than my previous post